Search results for: Generative Adversarial Attack
310 A Deep Reinforcement Learning-Based Secure Framework against Adversarial Attacks in Power System
Authors: Arshia Aflaki, Hadis Karimipour, Anik Islam
Abstract:
Generative Adversarial Attacks (GAAs) threaten critical sectors, ranging from fingerprint recognition to industrial control systems. Existing Deep Learning (DL) algorithms are not robust enough against this kind of cyber-attack. As one of the most critical industries in the world, the power grid is not an exception. In this study, a Deep Reinforcement Learning-based (DRL) framework assisting the DL model to improve the robustness of the model against GAAs is proposed. Real-world smart grid stability data, as an IIoT dataset, test our method and improve the classification accuracy of a DL model from around 57% to 96%.
Keywords: Generative Adversarial Attack, Deep Reinforcement Learning, deep learning, IIoT, Generative Adversarial Networks, power system.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 0309 Time Series Simulation by Conditional Generative Adversarial Net
Authors: Rao Fu, Jie Chen, Shutian Zeng, Yiping Zhuang, Agus Sudjianto
Abstract:
Generative Adversarial Net (GAN) has proved to be a powerful machine learning tool in image data analysis and generation. In this paper, we propose to use Conditional Generative Adversarial Net (CGAN) to learn and simulate time series data. The conditions include both categorical and continuous variables with different auxiliary information. Our simulation studies show that CGAN has the capability to learn different types of normal and heavy-tailed distributions, as well as dependent structures of different time series. It also has the capability to generate conditional predictive distributions consistent with training data distributions. We also provide an in-depth discussion on the rationale behind GAN and the neural networks as hierarchical splines to establish a clear connection with existing statistical methods of distribution generation. In practice, CGAN has a wide range of applications in market risk and counterparty risk analysis: it can be applied to learn historical data and generate scenarios for the calculation of Value-at-Risk (VaR) and Expected Shortfall (ES), and it can also predict the movement of the market risk factors. We present a real data analysis including a backtesting to demonstrate that CGAN can outperform Historical Simulation (HS), a popular method in market risk analysis to calculate VaR. CGAN can also be applied in economic time series modeling and forecasting. In this regard, we have included an example of hypothetical shock analysis for economic models and the generation of potential CCAR scenarios by CGAN at the end of the paper.
Keywords: Conditional Generative Adversarial Net, market and credit risk management, neural network, time series.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1215308 A Simulated Environment Approach to Investigate the Effect of Adversarial Perturbations on Traffic Sign for Automotive Software-in-Loop Testing
Authors: Sunil Patel, Pallab Maji
Abstract:
To study the effect of adversarial attack environment must be controlled. Autonomous driving includes mainly 5 phases sense, perceive, map, plan, and drive. Autonomous vehicles sense their surrounding with the help of different sensors like cameras, radars, and lidars. Deep learning techniques are considered Blackbox and found to be vulnerable to adversarial attacks. In this research, we study the effect of the various known adversarial attacks with the help of the Unreal Engine-based, high-fidelity, real-time raytraced simulated environment. The goal of this experiment is to find out if adversarial attacks work in moving vehicles and if an unknown network may be targeted. We discovered that the existing Blackbox and Whitebox attacks have varying effects on different traffic signs. We observed that attacks that impair detection in static scenarios do not have the same effect on moving vehicles. It was found that some adversarial attacks with hardly noticeable perturbations entirely blocked the recognition of certain traffic signs. We observed that the daylight condition has a substantial impact on the model's performance by simulating the interplay of light on traffic signs. Our findings have been found to closely resemble outcomes encountered in the real world.
Keywords: Adversarial attack simulation, computer simulation, ray-traced environment, realistic simulation, unreal engine.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 440307 Semi-Supervised Outlier Detection Using a Generative and Adversary Framework
Authors: Jindong Gu, Matthias Schubert, Volker Tresp
Abstract:
In many outlier detection tasks, only training data belonging to one class, i.e., the positive class, is available. The task is then to predict a new data point as belonging either to the positive class or to the negative class, in which case the data point is considered an outlier. For this task, we propose a novel corrupted Generative Adversarial Network (CorGAN). In the adversarial process of training CorGAN, the Generator generates outlier samples for the negative class, and the Discriminator is trained to distinguish the positive training data from the generated negative data. The proposed framework is evaluated using an image dataset and a real-world network intrusion dataset. Our outlier-detection method achieves state-of-the-art performance on both tasks.Keywords: Outlier detection, generative adversary networks, semi-supervised learning.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1080306 ECG-Based Heartbeat Classification Using Convolutional Neural Networks
Authors: Jacqueline R. T. Alipo-on, Francesca I. F. Escobar, Myles J. T. Tan, Hezerul Abdul Karim, Nouar AlDahoul
Abstract:
Electrocardiogram (ECG) signal analysis and processing are crucial in the diagnosis of cardiovascular diseases which are considered as one of the leading causes of mortality worldwide. However, the traditional rule-based analysis of large volumes of ECG data is time-consuming, labor-intensive, and prone to human errors. With the advancement of the programming paradigm, algorithms such as machine learning have been increasingly used to perform an analysis on the ECG signals. In this paper, various deep learning algorithms were adapted to classify five classes of heart beat types. The dataset used in this work is the synthetic MIT-Beth Israel Hospital (MIT-BIH) Arrhythmia dataset produced from generative adversarial networks (GANs). Various deep learning models such as ResNet-50 convolutional neural network (CNN), 1-D CNN, and long short-term memory (LSTM) were evaluated and compared. ResNet-50 was found to outperform other models in terms of recall and F1 score using a five-fold average score of 98.88% and 98.87%, respectively. 1-D CNN, on the other hand, was found to have the highest average precision of 98.93%.
Keywords: Heartbeat classification, convolutional neural network, electrocardiogram signals, ECG signals, generative adversarial networks, long short-term memory, LSTM, ResNet-50.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 202305 Adversarial Disentanglement Using Latent Classifier for Pose-Independent Representation
Authors: Hamed Alqahtani, Manolya Kavakli-Thorne
Abstract:
The large pose discrepancy is one of the critical challenges in face recognition during video surveillance. Due to the entanglement of pose attributes with identity information, the conventional approaches for pose-independent representation lack in providing quality results in recognizing largely posed faces. In this paper, we propose a practical approach to disentangle the pose attribute from the identity information followed by synthesis of a face using a classifier network in latent space. The proposed approach employs a modified generative adversarial network framework consisting of an encoder-decoder structure embedded with a classifier in manifold space for carrying out factorization on the latent encoding. It can be further generalized to other face and non-face attributes for real-life video frames containing faces with significant attribute variations. Experimental results and comparison with state of the art in the field prove that the learned representation of the proposed approach synthesizes more compelling perceptual images through a combination of adversarial and classification losses.Keywords: Video surveillance, disentanglement, face detection.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 614304 Resilient Machine Learning in the Nuclear Industry: Crack Detection as a Case Study
Authors: Anita Khadka, Gregory Epiphaniou, Carsten Maple
Abstract:
There is a dramatic surge in the adoption of Machine Learning (ML) techniques in many areas, including the nuclear industry (such as fault diagnosis and fuel management in nuclear power plants), autonomous systems (including self-driving vehicles), space systems (space debris recovery, for example), medical surgery, network intrusion detection, malware detection, to name a few. Artificial Intelligence (AI) has become a part of everyday modern human life. To date, the predominant focus has been developing underpinning ML algorithms that can improve accuracy, while factors such as resiliency and robustness of algorithms have been largely overlooked. If an adversarial attack is able to compromise the learning method or data, the consequences can be fatal, especially but not exclusively in safety-critical applications. In this paper, we present an in-depth analysis of five adversarial attacks and two defence methods on a crack detection ML model. Our analysis shows that it can be dangerous to adopt ML techniques without rigorous testing, since they may be vulnerable to adversarial attacks, especially in security-critical areas such as the nuclear industry. We observed that while the adopted defence methods can effectively defend against different attacks, none of them could protect against all five adversarial attacks entirely.
Keywords: Resilient Machine Learning, attacks, defences, nuclear industry, crack detection.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 513303 An Adversarial Construction of Instability Bounds in LIS Networks
Authors: Dimitrios Koukopoulos
Abstract:
In this work, we study the impact of dynamically changing link slowdowns on the stability properties of packetswitched networks under the Adversarial Queueing Theory framework. Especially, we consider the Adversarial, Quasi-Static Slowdown Queueing Theory model, where each link slowdown may take on values in the two-valued set of integers {1, D} with D > 1 which remain fixed for a long time, under a (w, ¤ü)-adversary. In this framework, we present an innovative systematic construction for the estimation of adversarial injection rate lower bounds, which, if exceeded, cause instability in networks that use the LIS (Longest-in- System) protocol for contention-resolution. In addition, we show that a network that uses the LIS protocol for contention-resolution may result in dropping its instability bound at injection rates ¤ü > 0 when the network size and the high slowdown D take large values. This is the best ever known instability lower bound for LIS networks.Keywords: Network stability, quality of service, adversarial queueing theory, greedy scheduling protocols.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1235302 Mechanisms of Internet Security Attacks
Abstract:
Internet security attack could endanger the privacy of World Wide Web users and the integrity of their data. The attack can be carried out on today's most secure systems- browsers, including Netscape Navigator and Microsoft Internet Explorer. There are too many types, methods and mechanisms of attack where new attack techniques and exploits are constantly being developed and discovered. In this paper, various types of internet security attack mechanisms are explored and it is pointed out that when different types of attacks are combined together, network security can suffer disastrous consequences.Keywords: DoS, internet attacks, router attack, security, trojan, virus, worm, XSS.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2115301 Attacks and Counter Measures in BST Overlay Structure of Peer-To-Peer System
Authors: Guruprasad Khataniar, Hitesh Tahbildar, Prakriti Prava Das
Abstract:
There are various overlay structures that provide efficient and scalable solutions for point and range query in a peer-topeer network. Overlay structure based on m-Binary Search Tree (BST) is one such popular technique. It deals with the division of the tree into different key intervals and then assigning the key intervals to a BST. The popularity of the BST makes this overlay structure vulnerable to different kinds of attacks. Here we present four such possible attacks namely index poisoning attack, eclipse attack, pollution attack and syn flooding attack. The functionality of BST is affected by these attacks. We also provide different security techniques that can be applied against these attacks.Keywords: BST, eclipse attack, index poisoning attack, pollution attack, syn flooding attack.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1629300 A Systematic Construction of Instability Bounds in LIS Networks
Authors: Dimitrios Koukopoulos
Abstract:
In this work, we study the impact of dynamically changing link slowdowns on the stability properties of packetswitched networks under the Adversarial Queueing Theory framework. Especially, we consider the Adversarial, Quasi-Static Slowdown Queueing Theory model, where each link slowdown may take on values in the two-valued set of integers {1, D} with D > 1 which remain fixed for a long time, under a (w, p)-adversary. In this framework, we present an innovative systematic construction for the estimation of adversarial injection rate lower bounds, which, if exceeded, cause instability in networks that use the LIS (Longest-in- System) protocol for contention-resolution. In addition, we show that a network that uses the LIS protocol for contention-resolution may result in dropping its instability bound at injection rates p > 0 when the network size and the high slowdown D take large values. This is the best ever known instability lower bound for LIS networks.
Keywords: Parallel computing, network stability, adversarial queuing theory, greedy scheduling protocols.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1419299 A Reasoning Method of Cyber-Attack Attribution Based on Threat Intelligence
Authors: Li Qiang, Yang Ze-Ming, Liu Bao-Xu, Jiang Zheng-Wei
Abstract:
With the increasing complexity of cyberspace security, the cyber-attack attribution has become an important challenge of the security protection systems. The difficult points of cyber-attack attribution were forced on the problems of huge data handling and key data missing. According to this situation, this paper presented a reasoning method of cyber-attack attribution based on threat intelligence. The method utilizes the intrusion kill chain model and Bayesian network to build attack chain and evidence chain of cyber-attack on threat intelligence platform through data calculation, analysis and reasoning. Then, we used a number of cyber-attack events which we have observed and analyzed to test the reasoning method and demo system, the result of testing indicates that the reasoning method can provide certain help in cyber-attack attribution.
Keywords: Reasoning, Bayesian networks, cyber-attack attribution, kill chain, threat intelligence.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2682298 A Distinguish Attack on COSvd Cipher
Authors: Mohammad Ali Orumiehchi ha, R. Mirghadri
Abstract:
The COSvd Ciphers has been proposed by Filiol and others (2004). It is a strengthened version of COS stream cipher family denoted COSvd that has been adopted for at least one commercial standard. We propose a distinguish attack on this version, and prove that, it is distinguishable from a random stream. In the COSvd Cipher used one S-Box (10×8) on the final part of cipher. We focus on S-Box and use weakness this S-Box for distinguish attack. In addition, found a leak on HNLL that the sub s-boxes don-t select uniformly. We use this property for an Improve distinguish attack.
Keywords: Stream cipher, COSvd cipher, distinguish attack, nonlinear feedback shift registers, chaotic layer.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1160297 Selective Forwarding Attack and Its Detection Algorithms: A Review
Authors: Sushil Sarwa, Rajeev Kumar
Abstract:
The wireless mesh networks (WMNs) are emerging technology in wireless networking as they can serve large scale high speed internet access. Due to its wireless multi-hop feature, wireless mesh network is prone to suffer from many attacks, such as denial of service attack (DoS). We consider a special case of DoS attack which is selective forwarding attack (a.k.a. gray hole attack). In such attack, a misbehaving mesh router selectively drops the packets it receives rom its predecessor mesh router. It is very hard to detect that packet loss is due to medium access collision, bad channel quality or because of selective forwarding attack. In this paper, we present a review of detection algorithms of selective forwarding attack and discuss their advantage & disadvantage. Finally we conclude this paper with open research issues and challenges.
Keywords: CAD algorithm, CHEMAS, selective forwarding attack, watchdog & pathrater, wireless mesh network.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2804296 Enhancing Hand Efficiency of Smart Glass Cleaning Robot through Generative Design Module
Authors: Pankaj Gupta, Amit Kumar Srivastava, Nitesh Pandey
Abstract:
This article explores the domain of generative design in order to enhance the development of robot designs for innovative and efficient maintenance approaches for tall buildings. This study aims to optimize the design of robotic hands by focusing on minimizing mass and volume while ensuring they can withstand the specified pressure with equal strength. The research procedure is structured and systematic. The purpose of optimization is to enhance the efficiency of the robot and reduce the manufacturing expenses. The project seeks to investigate the application of generative design in order to optimize products. Autodesk Fusion 360 offers the capability to immediately apply the generative design functionality to the solid model. The effort involved creating a solid model of the Smart Glass Cleaning Robot and optimizing one of its components, the Hand, using generative techniques. The article has thoroughly examined the designs, outcomes, and procedure. These loads serve as a benchmark for creating designs that can endure the necessary level of pressure and preserve their structural integrity. The efficacy of the generative design process is contingent upon the selection of materials, as different materials possess distinct physical attributes. The study utilizes five different materials, namely Steel, Stainless Steel, Titanium, Aluminum, and CFRP (Carbon Fiber Reinforced Polymer), in order to investigate a range of design possibilities.
Keywords: Generative design, mass and volume optimization, material strength analysis, generative design, smart glass cleaning robot.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 223295 How Efficiency of Password Attack Based on a Keyboard
Authors: Hsien-cheng Chou, Fei-pei Lai, Hung-chang Lee
Abstract:
At present, dictionary attack has been the basic tool for recovering key passwords. In order to avoid dictionary attack, users purposely choose another character strings as passwords. According to statistics, about 14% of users choose keys on a keyboard (Kkey, for short) as passwords. This paper develops a framework system to attack the password chosen from Kkeys and analyzes its efficiency. Within this system, we build up keyboard rules using the adjacent and parallel relationship among Kkeys and then use these Kkey rules to generate password databases by depth-first search method. According to the experiment results, we find the key space of databases derived from these Kkey rules that could be far smaller than the password databases generated within brute-force attack, thus effectively narrowing down the scope of attack research. Taking one general Kkey rule, the combinations in all printable characters (94 types) with Kkey adjacent and parallel relationship, as an example, the derived key space is about 240 smaller than those in brute-force attack. In addition, we demonstrate the method's practicality and value by successfully cracking the access password to UNIX and PC using the password databases createdKeywords: Brute-force attack, dictionary attack, depth-firstsearch, password attack.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 3481294 Vision Based Hand Gesture Recognition Using Generative and Discriminative Stochastic Models
Authors: Mahmoud Elmezain, Samar El-shinawy
Abstract:
Many approaches to pattern recognition are founded on probability theory, and can be broadly characterized as either generative or discriminative according to whether or not the distribution of the image features. Generative and discriminative models have very different characteristics, as well as complementary strengths and weaknesses. In this paper, we study these models to recognize the patterns of alphabet characters (A-Z) and numbers (0-9). To handle isolated pattern, generative model as Hidden Markov Model (HMM) and discriminative models like Conditional Random Field (CRF), Hidden Conditional Random Field (HCRF) and Latent-Dynamic Conditional Random Field (LDCRF) with different number of window size are applied on extracted pattern features. The gesture recognition rate is improved initially as the window size increase, but degrades as window size increase further. Experimental results show that the LDCRF is the best in terms of results than CRF, HCRF and HMM at window size equal 4. Additionally, our results show that; an overall recognition rates are 91.52%, 95.28%, 96.94% and 98.05% for CRF, HCRF, HMM and LDCRF respectively.
Keywords: Statistical Pattern Recognition, Generative Model, Discriminative Model, Human Computer Interaction.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2944293 An Atomic-Domains-Based Approach for Attack Graph Generation
Authors: Fangfang Chen, Chunlu Wang, Zhihong Tian, Shuyuan Jin, Tianle Zhang
Abstract:
Attack graph is an integral part of modeling the overview of network security. System administrators use attack graphs to determine how vulnerable their systems are and to determine what security measures to deploy to defend their systems. Previous methods on AGG(attack graphs generation) are aiming at the whole network, which makes the process of AGG complex and non-scalable. In this paper, we propose a new approach which is simple and scalable to AGG by decomposing the whole network into atomic domains. Each atomic domain represents a host with a specific privilege. Then the process for AGG is achieved by communications among all the atomic domains. Our approach simplifies the process of design for the whole network, and can gives the attack graphs including each attack path for each host, and when the network changes we just carry on the operations of corresponding atomic domains which makes the process of AGG scalable.Keywords: atomic domain, vulnerability, attack graphs, generation, computer security
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1660292 Graphical Password Security Evaluation by Fuzzy AHP
Authors: Arash Habibi Lashkari, Azizah Abdul Manaf, Maslin Masrom
Abstract:
In today's day and age, one of the important topics in information security is authentication. There are several alternatives to text-based authentication of which includes Graphical Password (GP) or Graphical User Authentication (GUA). These methods stems from the fact that humans recognized and remembers images better than alphanumerical text characters. This paper will focus on the security aspect of GP algorithms and what most researchers have been working on trying to define these security features and attributes. The goal of this study is to develop a fuzzy decision model that allows automatic selection of available GP algorithms by taking into considerations the subjective judgments of the decision makers who are more than 50 postgraduate students of computer science. The approach that is being proposed is based on the Fuzzy Analytic Hierarchy Process (FAHP) which determines the criteria weight as a linear formula.Keywords: Graphical Password, Authentication Security, Attack Patterns, Brute force attack, Dictionary attack, Guessing Attack, Spyware attack, Shoulder surfing attack, Social engineering Attack, Password Entropy, Password Space.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1942291 Effect of Amplitude and Mean Angle of Attack on Wake of an Oscillating Airfoil
Authors: Sadeghi H., Mani M., Ardakani M. A.
Abstract:
The unsteady wake of an EPPLER 361 airfoil in pitching motion has been investigated in a subsonic wind tunnel by hot-wire anemometry. The airfoil was given the pitching motion about the one-quarter chord axis at reduced frequency of 0182. Streamwise mean velocity profiles (wake profiles) were investigated at several vertically aligned points behind the airfoil at one-quarter chord downstream distance from trailing edge. Oscillation amplitude and mean angle of attack were varied to determine the effects on wake profiles. When the maximum dynamic angle of attack was below the static stall angle of attack, weak effects on wake were found by increasing oscillation amplitude and mean angle of attack. But, for higher angles of attack strong unsteady effects were appeared on the wake.Keywords: Unsteady wake, amplitude, mean angle, EPPLER 361 airfoil.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2685290 An Attack on the Lucas Based El-Gamal Cryptosystem in the Elliptic Curve Group Over Finite Field Using Greater Common Divisor
Authors: Lee Feng Koo, Tze Jin Wong, Pang Hung Yiu, Nik Mohd Asri Nik Long
Abstract:
Greater common divisor (GCD) attack is an attack that relies on the polynomial structure of the cryptosystem. This attack required two plaintexts differ from a fixed number and encrypted under same modulus. This paper reports a security reaction of Lucas Based El-Gamal Cryptosystem in the Elliptic Curve group over finite field under GCD attack. Lucas Based El-Gamal Cryptosystem in the Elliptic Curve group over finite field was exposed mathematically to the GCD attack using GCD and Dickson polynomial. The result shows that the cryptanalyst is able to get the plaintext without decryption by using GCD attack. Thus, the study concluded that it is highly perilous when two plaintexts have a slight difference from a fixed number in the same Elliptic curve group over finite field.
Keywords: Decryption, encryption, elliptic curve, greater common divisor.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 714289 Cryptographic Attack on Lucas Based Cryptosystems Using Chinese Remainder Theorem
Authors: Tze Jin Wong, Lee Feng Koo, Pang Hung Yiu
Abstract:
Lenstra’s attack uses Chinese remainder theorem as a tool and requires a faulty signature to be successful. This paper reports on the security responses of fourth and sixth order Lucas based (LUC4,6) cryptosystem under the Lenstra’s attack as compared to the other two Lucas based cryptosystems such as LUC and LUC3 cryptosystems. All the Lucas based cryptosystems were exposed mathematically to the Lenstra’s attack using Chinese Remainder Theorem and Dickson polynomial. Result shows that the possibility for successful Lenstra’s attack is less against LUC4,6 cryptosystem than LUC3 and LUC cryptosystems. Current study concludes that LUC4,6 cryptosystem is more secure than LUC and LUC3 cryptosystems in sustaining against Lenstra’s attack.Keywords: Lucas sequence, Dickson Polynomial, faulty signature, corresponding signature, congruence.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 738288 NANCY: Combining Adversarial Networks with Cycle-Consistency for Robust Multi-Modal Image Registration
Authors: Mirjana Ruppel, Rajendra Persad, Amit Bahl, Sanja Dogramadzi, Chris Melhuish, Lyndon Smith
Abstract:
Multimodal image registration is a profoundly complex task which is why deep learning has been used widely to address it in recent years. However, two main challenges remain: Firstly, the lack of ground truth data calls for an unsupervised learning approach, which leads to the second challenge of defining a feasible loss function that can compare two images of different modalities to judge their level of alignment. To avoid this issue altogether we implement a generative adversarial network consisting of two registration networks GAB, GBA and two discrimination networks DA, DB connected by spatial transformation layers. GAB learns to generate a deformation field which registers an image of the modality B to an image of the modality A. To do that, it uses the feedback of the discriminator DB which is learning to judge the quality of alignment of the registered image B. GBA and DA learn a mapping from modality A to modality B. Additionally, a cycle-consistency loss is implemented. For this, both registration networks are employed twice, therefore resulting in images ˆA, ˆB which were registered to ˜B, ˜A which were registered to the initial image pair A, B. Thus the resulting and initial images of the same modality can be easily compared. A dataset of liver CT and MRI was used to evaluate the quality of our approach and to compare it against learning and non-learning based registration algorithms. Our approach leads to dice scores of up to 0.80 ± 0.01 and is therefore comparable to and slightly more successful than algorithms like SimpleElastix and VoxelMorph.Keywords: Multimodal image registration, GAN, cycle consistency, deep learning.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 818287 A Novel Approach to Avoid Billing Attack on VOIP System
Authors: Narendra M. Shekokar, Satish R. Devane
Abstract:
In a recent year usage of VoIP subscription has increased tremendously as compare to Public Switching Telephone System(PSTN). A VoIP subscriber would like to know the exact tariffs of the calls made using VoIP. As the usage increases, the rate of fraud is also increases, causing users complain about excess billing. This in turn hampers the growth of VoIP .This paper describe the common frauds and attack on VoIP based system and make an attempt to solve the billing attack by creating secured channel between caller and callee.
Keywords: VoIP, Billing-fraud, SSL/TLS, MITM, Replay-attack.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1657286 Analysis of Detecting Wormhole Attack in Wireless Networks
Authors: Khin Sandar Win
Abstract:
In multi hop wireless systems, such as ad hoc and sensor networks, mobile ad hoc network applications are deployed, security emerges as a central requirement. A particularly devastating attack is known as the wormhole attack, where two or more malicious colluding nodes create a higher level virtual tunnel in the network, which is employed to transport packets between the tunnel end points. These tunnels emulate shorter links in the network. In which adversary records transmitted packets at one location in the network, tunnels them to another location, and retransmits them into the network. The wormhole attack is possible even if the attacker has not compromised any hosts and even if all communication provides authenticity and confidentiality. In this paper, we analyze wormhole attack nature in ad hoc and sensor networks and existing methods of the defending mechanism to detect wormhole attacks without require any specialized hardware. This analysis able to provide in establishing a method to reduce the rate of refresh time and the response time to become more faster.Keywords: Ad hoc network, Sensor network, Wormhole attack, defending mechanism.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2307285 Security Weaknesses of Dynamic ID-based Remote User Authentication Protocol
Authors: Hyoungseob Lee, Donghyun Choi, Yunho Lee, Dongho Won, Seungjoo Kim
Abstract:
Recently, with the appearance of smart cards, many user authentication protocols using smart card have been proposed to mitigate the vulnerabilities in user authentication process. In 2004, Das et al. proposed a ID-based user authentication protocol that is secure against ID-theft and replay attack using smart card. In 2009, Wang et al. showed that Das et al.-s protocol is not secure to randomly chosen password attack and impersonation attack, and proposed an improved protocol. Their protocol provided mutual authentication and efficient password management. In this paper, we analyze the security weaknesses and point out the vulnerabilities of Wang et al.-s protocol.Keywords: Message Alteration Attack, Impersonation Attack
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1779284 A New Traffic Pattern Matching for DDoS Traceback Using Independent Component Analysis
Authors: Yuji Waizumi, Tohru Sato, Yoshiaki Nemoto
Abstract:
Recently, Denial of Service(DoS) attacks and Distributed DoS(DDoS) attacks which are stronger form of DoS attacks from plural hosts have become security threats on the Internet. It is important to identify the attack source and to block attack traffic as one of the measures against these attacks. In general, it is difficult to identify them because information about the attack source is falsified. Therefore a method of identifying the attack source by tracing the route of the attack traffic is necessary. A traceback method which uses traffic patterns, using changes in the number of packets over time as criteria for the attack traceback has been proposed. The traceback method using the traffic patterns can trace the attack by matching the shapes of input traffic patterns and the shape of output traffic pattern observed at a network branch point such as a router. The traffic pattern is a shapes of traffic and unfalsifiable information. The proposed trace methods proposed till date cannot obtain enough tracing accuracy, because they directly use traffic patterns which are influenced by non-attack traffics. In this paper, a new traffic pattern matching method using Independent Component Analysis(ICA) is proposed.
Keywords: Distributed Denial of Service, Independent Component Analysis, Traffic pattern
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1777283 A New Knapsack Public-Key Cryptosystem Based on Permutation Combination Algorithm
Authors: Min-Shiang Hwang, Cheng-Chi Lee, Shiang-Feng Tzeng
Abstract:
A new secure knapsack cryptosystem based on the Merkle-Hellman public key cryptosystem will be proposed in this paper. Although it is common sense that when the density is low, the knapsack cryptosystem turns vulnerable to the low-density attack. The density d of a secure knapsack cryptosystem must be larger than 0.9408 to avoid low-density attack. In this paper, we investigate a new Permutation Combination Algorithm. By exploiting this algorithm, we shall propose a novel knapsack public-key cryptosystem. Our proposed scheme can enjoy a high density to avoid the low-density attack. The density d can also exceed 0.9408 to avoid the low-density attack.Keywords: Public key, Knapsack problem, Knapsack cryptosystem, low-density attack.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1963282 Studies on the Blended Concrete Prepared with Tannery Effluent
Authors: K. Nirmalkumar
Abstract:
There is a acute water problem especially in the dry season in and around Perundurai (Erode district, Tamil Nadu, India) where there are more number of tannery units. Hence an attempt was made to use the waste water from tannery industry for construction purpose. The mechanical properties such as compressive strength, tensile strength, flexural strength etc were studied by casting various concrete specimens in form of cube, cylinders and beams etc and were found to be satisfactory. Hence some special properties such as chloride attack, sulphate attack and chemical attack are considered and comparatively studied with the conventional potable water. In this experimental study the results of specimens prepared by using treated and untreated tannery effluent were compared with the concrete specimens prepared by using potable water. It was observed that the concrete had some reduction in strength while subjected to chloride attack, sulphate attack and chemical attack. So admixtures were selected and optimized in suitable proportion to counter act the adverse effects and the results were found to be satisfactory.Keywords: Calcium nitrite, concrete, fly ash.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1999281 On Dialogue Systems Based on Deep Learning
Authors: Yifan Fan, Xudong Luo, Pingping Lin
Abstract:
Nowadays, dialogue systems increasingly become the way for humans to access many computer systems. So, humans can interact with computers in natural language. A dialogue system consists of three parts: understanding what humans say in natural language, managing dialogue, and generating responses in natural language. In this paper, we survey deep learning based methods for dialogue management, response generation and dialogue evaluation. Specifically, these methods are based on neural network, long short-term memory network, deep reinforcement learning, pre-training and generative adversarial network. We compare these methods and point out the further research directions.Keywords: Dialogue management, response generation, reinforcement learning, deep learning, evaluation.
Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 792